The overarching goal of the FBIRN is to develop technology and methods to conduct multisite functional imaging studies, and to produce a knowledge base that would not otherwise be available through single-site imaging studies. The technology includes the development and refinement of multi-site, functional imaging protocols using robust cognitive tasks; the development and refinement of algorithms to reduce inter-site variability; the use of federated, distributed databases and storage; flexible and robust image processing software integrated with these databases and storage infrastructure; tools for data querying across the distributed databases to extract subsets of data from multiple sources; and the development of both classical statistical and data mining methods to reveal the patterns of imaging, clinical, and behavioral data which differentiate important population clusters. To meet this goal, the FBIRN will develop the capacity to conduct a multi-center functional imaging study in a focused group of subjects with a neuropsychiatric disorder, patients with schizophrenia. The lessons learned, the statistical methods developed, and the informatics structure constructed will be generalizable and applicable to a wide variety of clinical investigations. The ability to integrate subsets of clinical, functional imaging, and behavioral datasets from disparate sites to form novel datasets will be assessed in a phased series of developments and optimizations. The overall goal of this proposal is to develop tools to make multi-site fMRI studies a common research practice. Completion of this goal enables researchers to investigate the pathophysiology of complex diseases more thoroughly, through the increased power of large-scale, collaborative neuroimaging studies. When completed, researchers in different fields and physical locations will be able to draw on a common set of tools and database systems not possible at a single site.